Questions tagged [sample-size]
This tag is ambiguous. Use it when the question is about sample size and NONE of the following are more appropriate: [small-sample], [large-data], [statistical-power], [underdetermined], or [unbalanced-classes].
2,019 questions
1
vote
0
answers
77
views
Bias in standard error of regression slope with not-independent data and effective sample size
Consider a sample of $N/2$ pairs of individuals. Each pair belongs to a group $j$.
For each individual $i$ from the $N$ sample, I measure two variables ($y_{i}$ and $x_{i}$) and the average per group $...
- 267
5
votes
2
answers
376
views
Sample size calculation for a test on proportions
I'm not sure whether I'm assessing this problem correctly.
Suppose I want do demonstrate that an adverse event occurs in less than 20% of the participants (but I hope/assume that it actually NEVER ...
- 123
7
votes
1
answer
192
views
Doubling your accuracy - extension
Frederick Mosteller's 50 Challenging Problems in Probability has a nice question I have not seen before, and I was wondering whether it could be extended.
49. Doubling your accuracy
An unbiased ...
- 45.2k
0
votes
0
answers
41
views
Estimate probability distribution of pretrained language model
Suppose you have a pretrained protein language model like and you can let it generate some sequences of a given length L. Given a number of such samples, you want to estimate the probability ...
- 101
3
votes
1
answer
130
views
Sample size for multilevel exploratory factor analysis
I have a dataset of n=90 participants with ~20 variables for two conditions, with each variable having 4 data points (not statistically independent) for each condition. I am thinking about conducting ...
- 33
4
votes
3
answers
172
views
How to be confident of some program failing less often than a threshold?
When writing complex software, it is common for some of the tests to be "flaky", failing in a seemingly random fashion some fraction of the time. Such failures can often be eliminated by ...
- 183
5
votes
1
answer
386
views
How many samples do I need to decide if there is a tie?
There are some 1,000,000 voters who vote for one of two candidates, A or B. I would like to decide if the election outcome would be a tie (that is, 500,000 vote for A and 500,000 vote for B). I poll ...
- 1,091
4
votes
1
answer
68
views
Sample size calculation Split Plot design
I'm currently trying to calculate the sample size for a completely randomized split-plot design. The design includes two treatments at the whole-plot level and two treatments at the sub-plot level. ...
- 41
0
votes
0
answers
50
views
Naïve sample size calculation of non inferiority mixed model
First of all, thank you for this wonderful forum! I’m a physician with some statistics background but not a professional statistician. We plan an academic non-inferiority retrospective study with a ...
- 31
1
vote
0
answers
66
views
Reducing sample size for an equality of opportunity test
We have created a dataset of a few hundred items that gives us two sets of scores in the range $[0,1]$. The scores do not follow a well known distribution and is multi-modal. I want to test that these ...
- 11
6
votes
2
answers
153
views
Are there rules of thumb for the sample size required when using a categorical predictor in linear regression?
I’ve had a reviewer suggest that I use ethnicity as a covariate in a linear regression. Some ethnic groups in the sample are small enough that I am a little worried that I will overfit if I do this.
...
- 1,091
0
votes
0
answers
55
views
Under what conditions is penalised regression appropriate?
In this question posted earlier this year I asked about strange results from a penalised regression regression model: stacked elastic net regression in fact. The CV member who answered my question ...
- 2,293
3
votes
1
answer
89
views
Sample size for (average) bioequivalence in R
The aim id to calculate the sample size for average bioequivalence trial.
I would like to replicate the below example from "Sample Size Calculations in Clinical Research" by Chow and Shao (...
- 131
0
votes
0
answers
42
views
Power and sample size, original sample size estimation was 34, current sample size is 13 and power is above .8 [duplicate]
I have been doing supervised UG research for the last year, and when I did an initial power analysis for the experimental phase in January (preliminary phase was questionnaire data and recruitment) a ...
1
vote
0
answers
68
views
How can I shorten an A/B test on a binary registration metric when CUPED isn’t possible (new users only)?
Context
────────
• Metric: registration_success (binary 0/1, first-time users only)
• Baseline conversion rate: 15 %
• Target lift (MDE): +20 % relative (18 % vs 15 %)
• Power / α: 80 %, two-sided 5 %
...
